OVARIAN INSUFFICIENCY

CROSS-REFERENCE TO RELATED APPLICATIONS

This patent appl ication claims priority from Italian patent application no . 102021000002228 filed on February 2 , 2021 , the entire disclosure of which is incorporated herein by reference .

TECHNICAL SECTOR

The present invention relates to a composition based on amino acids , in particular for use as a food supplement to limit metabolic imbalances in the liver resulting from insuf ficient ovarian estrogen production due to age ( climacteric ) or to congenital or acquired hypogonadism .

STATE OF THE PRIOR ART

Throughout phylogeny, the liver, or its ancestral counterpart organ, has taken on a primary role in the control of reproduction as it is responsible for the synthesis of proteins essential for egg maturation; in turn, the gonads regulate several hepatic metabolic functions connected to reproductive needs .

This close , reciprocal control is strictly maintained throughout evolution from oviparous to viviparous species . However, plausibly, gonadal control of hepatic metabolism increased signi ficantly in the transition from oviparous to mammals to allow the female liver to perceive and respond to a greater number of stresses connected to the dramatic changes connected to the di f ferent stages of reproduction ( follicle growth and maturation, pregnancy, lactation, menopause ) .

As proof of this hypothesis correctness , the liver, as happens for the brain, undergoes a sexual di f ferentiation by oestrogens and is the organ with the highest degree of sexual dimorphism .

The primary mammalian hepatic sensor of gonadal function is the estrogen receptor alpha (Era) isoform . In the liver of female mammals , ERa regulates hepatic metabolism based on the reproductive state and this is possible as each of these states is characterised by di f ferent concentrations of circulating oestrogens (El , E2 , E3 , E4 ) . For example , during the reproductive cycle , the high plasma estrogen content typical of the pre-ovulatory phase is connected with the blockade of the de novo lipid synthesis and with the changes in lipid transport , with an increased clearance of LDL ( low density lipoprotein) and VLDL (very low density lipoprotein) , of the synthesis of HDL (high density lipoprotein) , of the changes in lipoprotein modi fiers and hepatic lipoprotein receptors . The physiological relevance of oestrogens and ERa in all these changes is demonstrated by the fact that they are cancelled by selective ablation of hepatic ERa ( LERKO) .

In order to perform this complexity of functions , the hepatic ERa is able to integrate a multiplicity of signals (hormonal and non-hormonal ) that allow this powerful transcription factor to select and modulate , among the numerous potential target genes , those that best support the energy requirements of that speci fic reproductive state . It is now known that , in addition to oestrogens , ERa can be activated by various signal molecules such as growth factors (EGF, IGF) or amino acids . In particular, the activation of the hepatic ERa by amino acids in the diet is neces sary for the proper progression of the oestrous/menstrual cycle and represents a well-preserved element in evolution because it aims to regulate fertility according to the availability of essential nutrients such as amino acids .

In case of insuf ficient nutritional intake, the reduced activation of hepatic ERa leads to a reduction in the synthesis and secretion of factors relevant to the progression of the oestrous cycle ( such as IGF1 ) and leads to a progressive blockage of the reproductive cycle , thus preventing reproduction under unfavourable energy conditions . As evidenced by studies conducted using di f ferent food regimes such as fasting or a fat-enriched diet (HFD) , this sex-speci fic mechanism underlies a strong hepatic sexual dimorphism in amino acid metabolism .

In case of short-term fasting, the female liver catabolises the amino acids to fuel lipid synthesis by generating NADPH through the pentose phosphate pathway; this metabolic adaptation does not occur in LERKO females , who do not express ERa in the liver, as well as in males , further underlining how this ERa-dependent regulatory mechanism is a prerogative of females .

In case of prolonged exposure to HFD, females retain amino acid homeostasis and do not accumulate lipids in the liver ; this is not the case for LERKO females and males ( for the latter regardless of the presence or absence of hepatic ERa) , in whose livers the fat-enriched diet induces signi ficant lipid deposition and a decrease in the content of essential amino acids (EAA) and branched chain amino acids (BCAA) .

The combination of these experiments indicates that , in mammalian females , hepatic ERa plays an important role in the regulation of lipid metabolism and amino acid homeostasis .

Considering that in female mammals , including humans , the cessation of the ovarian functions has serious cardio- metabolic consequences and that the treatments for ovarian dys functions , especially long-term treatments , have detrimental ef fects on health, there is a pressing need in the art to identi fy alternative treatments to the administration of oestrogens .

OBJECT OF THE INVENTION

Aim of the present invention is therefore to provide a composition useful as a food supplement capable of stimulating hepatic ERa activity in females with low or altered levels of oestrogens , as found in menopause and hypogonadal conditions .

This aim is achieved by means of a composition according to Claim 1 . BRIEF DESCRIPTION OF THE DRAWINGS

To better understand the present invention preferred embodiments thereof will be now described, for exemplary and non-limiting purposes, with reference to the appended claims, wherein:

- Figure 1 shows, as Log2 C, the expression of genes altered by ovariectomy (OVX) in the liver of female ERa ^f/f mice (control animals expressing the hepatic estrogen receptor) . These genes are connected to the lipid biosynthetic process (panel a) , the acyl-CoA metabolism (panel b) , the p-oxidation of fatty acids (panel c) , electron transport coupled to mitochondrial ATP synthesis (panel d) , and the Wnt signalling pathway (panel e) obtained by RNA-Seq analysis (n=4) ;

- Figure 2 shows the number of differentially expressed genes between OVX mice and SHAM mice (operated in a faked way) depending on the diet given: the number of genes modulated by ovariectomy is much lower when the diet enriched in essential amino acids is given;

- Figure 3 shows the number of genes up- and down- regulated by CRTL and *AA diets;

- Figure 4 shows (a, e) the experimental scheme adopted to evaluate the effects of the *AA diet in ERa ^f/f (a) and LERKO (e) OVX females; (b, f) the body weight (BW) of SHAM and OVX ERa ^f/f (b) and LERKO (f) female mice fed the CTRL or *AA diet measured weekly for 12 weeks and expressed as a percentage compared to time 0. The data are the mean ± SEM (n=8) . *p<0.05 and ***p<0.001 OVX vs SHAM; °p<0.05 LERKO vs ERa ^f/f with two-way ANOVA followed by Bonferroni's post hoc test; (c, g) representative images of lipid deposits by Oil Red 0 staining in liver tissues of SHAM and OVX ERa ^f/f (c) and LERKO (g) female mice fed CTRL or *AA diet for 12 weeks. Magnification x200; (d, h) the quantification of Oil Red 0 staining shown in the panels b and e (d is relative to c; h is relative to g) . The data are expressed as percentages of the total areas of the section. The data are the mean ± SEM (n=6) . *p<0.05 and ***p<0.001 OVX vs SHAM; °p<0.05 LERKO vs ERa ^f/f with two-way ANOVA followed by Bonferroni's post hoc test .

Figure 5 shows (a) the body weight (expressed as % relative to time 0) of SHAM and OVX ERa ^f/f and LERKO females fed the CTRL or *AA diet measured at the end of the 12-week experiment; (b) food intake expressed as g/day/mouse; (c) feed efficiency expressed as ABW/AFI (Abody weight/Afood intake) ; (d) leptin levels measured in the plasma of SHAM and OVX ERa ^f/f (a) and LERKO (b) females fed the CTRL or *AA diet. The data are shown as mean ± SEM (n = 8) . *p<0.05, **p<0.01 and ***p<0.001 OVX vs SHAM; °p<0.05 and ^{o o o} p<0.001 LERKO vs ERa ^f/f by two-way ANOVA followed by Bonferroni's post hoc test.

Figure 6 shows (a) the experimental scheme aimed at defining the involvement of hepatic ERa in the response to the *AA diet; (b) Venn diagram to compare the response of ERa ^f/f females to the *AA diet. The differentially expressed genes (DEG) from the RNA-Seq analysis (n=4) were considered significant when | FC | >1.5 and padj<0.1; (c) the distribution of genes up- and down-regulated by the *AA diet in the liver of ERa ^f/f and LERKO females by RNA-Seq analysis (n=4) ; (d-e) Volcano plot of the DEGs identified by RNA-Seq (n=4) in the liver of ERa ^f/f (d) and LERKO (e) females fed the CTRL or *AA diet. The *AA/CTRL ratio of gene expression is shown on the

X-axis as Log2FC and the significance is displayed on the Y- axis as -Logiopadj . The genes significantly up- and down- regulated (with | FC | >1.5; padj<0.1) by the *AA diet are in light and dark grey, respectively; (f-i) gene ontology (GO) analysis generated with REVIGO. The scatterplot of GO terms obtained with Enrichr were correlated to the genes up- (f) and down-regulated (g) by the *AA diet exclusively in the liver of ERa ^f/f and to the genes up- (h) and down-regulated (i) by the *AA diet exclusively in the liver of LERKO females. The dimension of each bubble reflects the count of each term in the list of enriched terms.

Figure 7 shows (a-d) the heatmaps reporting as Log2FC the expression of the genes modulated by the *AA diet exclusively in the liver of ERa ^f/f females and connected to ATP synthesis/metabolism and mitochondrial respiratory chain (a) , the p-catenin-TCF assembly pathway (b) , the cell cycle (c) , the transcription regulation (d) by RNA-Seq analysis (n=4) . (e-h) heatmaps reporting as Log2FC the expression of the genes modulated by the *AA diet exclusively in the liver of LERKO females and connected to DNA repair (e) , the TGF- pl signalling pathway (f) , the collagen metabolism (g) , the extracellular matrix organisation (h) by RNA-Seq analysis (n=4) .

Figure 8 shows the cluster analysis performed with the Genesis software to identify the number of genes belonging to the "restored", "unchanged" and "divergent" class in ERa ^f/f and LERKO females. Figure 9 shows ( a ) the Venn diagram summarising the overlap between the di f ferentially expressed genes ( DEG) identi fied by RNA-Seq (n=4 ) in the liver of male, fertile female mice and OVX ERa ^{f/ f} female mice ; (b-d) the heatmaps reporting as Log2FC the mean expression of the most enriched classes of di f ferentially expressed genes in the liver of OVX ERa ^{f/ f} females and in the liver of ERa ^{f/ f} males compared to fertile ERa ^{f/ f} females (n=4 ) .

PREFERRED EMBODIMENT OF THE INVENTION

According to a first aspect of the invention, there is therefore provided a composition for use as a food supplement to reduce hepatic metabolic imbalances resulting from insuf ficient ovarian estrogen production comprising at least the essential amino acids isoleucine, leucine , valine . The composition further comprises at least one essential amino acid selected from the group consisting of histidine , lysine , methionine , phenylalanine , tryptophan and threonine .

Surprisingly, the inventors have found that a diet enriched in essential amino acids is able to counteract the metabolic imbalance resulting from decreased plasma estrogen levels in a menopausal model , largely restoring a more proper regulation of the liver transcriptome and thus limiting the lipid deposit in the liver and the body weight gain .

In particular, an intake of at least 25-45% more, preferably between 50% and 120% , even more preferably 50-70% or 80- 120% more , than the daily requirement in essential amino acids has been found to be able to counteract the metabolic imbalances connected to ovarian insufficiency . The daily doses per kilo of weight recommended for daily intake for men are: histidine 13 mg/kg/day isoleucine 39 mg/kg/day leucine 78 mg/kg/day lysine 48 mg/kg/day methionine 9 mg/kg/kg valine 41 mg/kg/day threonine 26 mg/kg/day tryptophan 4 mg/kg/day According to the present invention, the composition comprises at least three essential amino acids selected from the following ones which are administered at a dosage of: 10.7-16 mg/kg/day for histidine, 31-47 mg/kg/day for isoleucine, 63-94 mg/kg/day for leucine, 38-57 mg/kg/day for lysine, 7.5-11.5 mg/kg/day for methionine, 13-20 mg/kg/day for phenylalanine, 3-5 mg/kg/day for tryptophan, 21-31 mg/kg/day for threonine, 33-49 mg/kg/day for valine. More preferably, the composition comprises at least three essential amino acids which are administered at a dosage of:

11.3-15 mg/kg/day for histidine, 33-45 mg/kg/day for isoleucine, 67-90 mg/kg/day for leucine, 41-55 mg/kg/day for lysine,

8-11 mg/kg/day for methionine,

14-19 mg/kg/day for phenylalanine,

3.3-4.5 mg/kg/day for tryptophan,

22-30 mg/kg/day for threonine,

35-47 mg/kg/day for valine.

Even more preferably, the at least three essential amino acids are administered at a dosage of:

12-14.7 mg/kg/day for histidine,

35-42 mg/kg/day for isoleucine,

71-86 mg/kg/day for leucine,

43-52 mg/kg/day for lysine,

8.5-10.5 mg/kg/day for methionine,

15-18 mg/kg/day for phenylalanine,

3.5-4.3 mg/kg/day for tryptophan,

23-29 mg/kg/day for threonine,

37-45 mg/kg/day for valine.

In a further embodiment, the composition comprises at least three amino acids selected from:

11.3-15 mg/kg/day for histidine,

33-45 mg/kg/day for isoleucine,

67-90 mg/kg/day for leucine,

41-55 mg/kg/day for lysine,

22-30 mg/kg/day for threonine,

33-49 mg/kg/day for valine; preferably :

12-14.7 mg/kg/day for histidine,

35-42 mg/kg/day for isoleucine, 71-86 mg/kg/day for leucine,

43-52 mg/kg/day for lysine,

23-29 mg/kg/day for threonine,

37-45 mg/kg/day for valine.

The composition of the invention may be used as a dietary supplement or food supplement, in particular to reduce hepatic metabolic imbalances, especially the lipid deposit in the liver and/or weight gain, resulting from ovarian insufficiency and thus to adjuvant the treatment of post-menopausal syndrome and in other metabolic disorders resulting from female hypogonadism.

Furthermore, the composition can be administered in addition to or in combination with estrogen-based therapies.

Further characteristics of the present invention will become apparent from the following description of some merely illustrative and non-limiting examples.

EXAMPLE

Metabolic effects of the composition of the invention in subjects undergoing ovariectomy

ERa ^f/f mice (mean weight 24.5 g) were ovariectomised (OVX) or sham-operated (SHAM) at 2 months of age; from month 3 onwards, the mice were fed for 12 weeks on a control diet (CRTL) or an iso-caloric/ iso-protein diet which provides, in addition to the control diet, also the intake of the composition of the invention (*AA) (Figure 4a) .

In particular, the two diets administered have an overall composition as shown in Table 1. Table 1

As regards the protein fraction, the two diets administered are represented in Table 2 .

Table 2

At the end of the dietary treatment, the mice were sacrificed by euthanasia after 6 hours of fasting. SHAM females were sacrificed in oestrus, the phase of the oestrous cycle characterised by low estrogen levels.

Transcriptomic analysis, performed by RNA-Seq, showed that in OVX mice important metabolic changes occurred in the liver with a significant increase in the expression of mRNAs encoding proteins for lipid biosynthetic processes (Figure la) , for acyl-CoA metabolism (Figure lb) , for p-oxidation of fatty acids (Figure 1c) , for mitochondrial ATP synthesis (Figure Id) , and a decrease in mRNAs of the Wnt signalling pathway (Figure le) . These changes are in line with what has been observed in the literature and have indicated a shift towards lipid anabolism (Della Torre et al., 2014; Paquette et al., 2008; Villa et al., 2012; Zhu et al., 2013) . The decrease in mRNAs encoding the members of the P450 protein family suggested that ovariectomy is also connected to a decreased detoxification power in the liver.

Considering a change in expression | FC | >1.5 and an adjusted value p padj<0.1, the hepatic transcriptome of the SHAM and OVX mice was compared: in the mice fed the CTRL diet the number of differentially expressed genes (DEG) was 912, but in the mice fed the *AA diet the number of DEGs was reduced by nearly 2/3 (308) (Figure 2) . The percentage of genes that were up- or down-regulated did not change significantly between the two diets (50% were up-regulated genes with CTRL and 60% with *AA) (Figure 3) .

Analysing the data using the Volcano plot, it was found that for several genes that remained differentially expressed even with the *AA diet, the extent of the difference (Log2 C) was significantly reduced (e.g. Cyp3al6 , cytochrome P450 3A16; Cyp4al2a, cytochrome P450 4A12A) . This led to a further analysis to be carried out on the effects of the diet in OVX mice with the application of the Genesis software, which allowed the DEGs to be grouped into three groups: genes whose expression was altered by ovariectomy and brought back to the levels of the SHAM mice by the *AA diet ("restored") ; genes modulated by ovariectomy but not modified by the *AA diet ("unchanged") ; genes, whose expression was not altered or modified by ovariectomy but was altered by the *AA diet ("divergent") .

In the group of the "restored" genes, the most significant classes are constituted by genes encoding proteins involved in the regulation of lipid metabolic processes, in lipid biosynthesis and metabolism, together with the regulation of hormone levels and the Wnt signalling pathway. In particular, it was observed that the *AA diet is able to fully or at least partially restore the level of expression of families of genes relevant mainly for detoxification (P450s and sulfotransferases) , transport and control of hepatic metabolism (Mups, major urinary proteins) . Beyond that, the *AA diet alleviated the signi ficant inhibition caused by OVX of several players in the Wnt signalling pathway .

The genes "unchanged" by the *AA diet are linked to biological KEGG processes identi fying oxidative phosphorylation, and acyl-CoA and fatty acid metabolic processes .

The "divergent" group comprises genes encoding proteins classi fied as regulators o f the immune responses and, to a lesser extent , proteins involved in fatty acid and cholesterol metabolism .

It should be noted that the maj ority of these mRNAs decreased following ovariectomy, but were more abundant with the diet than in the controls , suggesting that the diet , in an attempt to restore the ef fect of ovariectomy, exceeded the state of physiological normality ( this concerned in particular several mRNAs involved in the regulation of fatty acid and cholesterol metabolism and encoding proteins of inflammatory and apoptotic responses ) .

Taken together, these results lead to the conclusion that the composition of the invention ( *AA) plays a beneficial role in that it opposes the metabolic alterations following ovariectomy and alleviates the harmful active processes occurring in the liver of OVX mice , plausibly as a consequence of a decrease in detoxi fication and of an increase in oxidative processes .

The body weight (BW) and the lipid content in the liver were assessed to further test the extent to which changes in gene expression observed in OVX mice are connected to an overall effect of the *AA diet on the metabolism of the mice.

BW was measured weekly during exposure to the CTRL and *AA diets; as expected, over the 3 months of the experiment, the OVX mice fed the CTRL diet gained much more weight than SHAM mice ( + 13%) (Figure 4b and 5a) . This was not the case for the mice fed the *AA diet, for which, at the end of the treatment, weight gain was not significantly different from SHAM (Figure 4b and 5a) . The effects described above are in accordance with the changes in feed efficiency (FE) , whose ovariectomy-induced increase was mitigated by the *AA diet (Figure 5c) . Given the fact that the mice fed the CTRL and *AA diet ate a comparable amount of food (Figure 5b) , the hypothesis that the reduced BW of the ERa ^f/f OVX mice fed the *AA diet could be due to a decreased food intake was ruled out. Conversely, the increase in feed efficiency (FE) induced by OVX was mitigated by the *AA diet (Figure 5c) . This observation further supports the idea of a different regulation of metabolism in OVX females fed the *AA diet compared to those fed the CTRL diet.

Subsequently, the presence of lipid deposits in the liver was assessed. Staining with red oil 0 (specific for lipid detection) indicates that there is a significant increase (+41%) in lipid deposits in the liver parenchyma in the OVX mice fed the CTRL diet, but not the *AA diet (+9%) (Figure 4 c-d) .

These results indicated that the diet of the invention, enriched in essential amino acids (EAA) and especially in branched chain amino acids (BCAA) , is able to largely prevent the hepatic lipid metabolic imbalance induced by estrogen deficiency and prevented the formation of lipid deposits as suggested by transcriptomic data . Apart from the molecular events , the most relevant data is represented by the fact that the *AA diet mitigates the lipid deposit in the liver and body weight gain following ovariectomy .

The above-reported data demonstrate that amino acids are essential for the full activation of the hepatic ERa ( Della Torre et al . , 2011a ) and thus for the regulation of those liver functions that are under the control of this receptor in the female liver .

Only the availability of these new data allowed and led the inventors to hypothesise that in OVX mice , the *AA diet can compensate for the lack of oestrogens by restoring most of the transcriptional activity of hepatic ERa that is indispensable for the lipid homeostasis characteristic of the fertile period in female mammals .

To test to what extent hepatic ERa was involved in the response to the *AA diet , experiments were performed on LERKO ( Liver ERa knockout - Della Torre et al . , 2011a ) mice : LERKO females were subj ected to surgical or simulated ovariectomy ( SHAM) and then fed either the CTRL or *AA diet . Similarly to ERa ^{f/ f} mice , the ovariectomy induced a signi ficant increase in the BW of LERKO mice fed the CRTL diet , which weighed 13% more than their SHAM counterparts at the end of the dietary treatment . However, in LERKO OVX mice the *AA diet failed to completely counteract weight gain (BW) and feed ef ficiency

( EE ) ( Figure 4 f and 5a, c ) . In line with these results , the staining of the liver with red oil 0 showed that the *AA diet - unlike to what happens in the ERa ^{f/ f} mice - is not able to limit the lipid deposit in the liver of the LERKO OVX females ( Figure 4g-h) .

Finally, plasma leptin was measured in all the above experimental groups . Figure 5d shows that ovariectomy is connected to an increase in leptin levels in both ERa ^{f/ f} and LERKO mice , but the *AA diet had no influence on the secretion of this hormone .

Taken together, these results indicate that the hepatic ERa is essential in order to allow an *AA diet to re-establish in OVX females the liver metabolism that characterises fertile females , limiting metabolic alterations such as lipid deposit in the liver and body weight gain .

The role of the hepatic ERa in the response to the *AA diet

Finally, to better understand the metabolic role played by the hepatic ERa in the response to the *AA diet , the di f ferences in the hepatic transcriptome of ERa ^{f/ f} and LERKO mice after administration of the *AA diet were investigated ( Figure 6a ) .

Considering a | FC | >1 . 5 and a padj <0 . 1 , in ERa ^{f/ f} mice the *AA diet induced changes in the hepatic expression of 539 genes ( 311 up-regulated and 228 down-regulated) ( Figure 6b-c ) ; the response of LERKO mice to the *AA diet was more limited, af fecting only 296 genes , of which 195 were up- regulated and 101 down-regulated ( Figure 6b-c ) . The response to the *AA diet was also qualitatively different in LERKO mice, in that only 10% of the modulated genes were found to be in common with ERa ^f/f mice (Figure 6b) . Furthermore, Volcano plots showed that in ERa ^f/f mice the changes induced by the *AA diet are very extensive, particularly in the case of up-regulated genes; indeed, after the *AA diet the expression of several genes changed by several orders of magnitude with a very high level of significance (Figure 6d) .

The most affected genes were Lars2 (leucyl-tRNA synthetase 2) , which encodes an enzyme involved in tRNA aminoacylation, and several major urinary proteins (i.e. Mup7, Mupll, Mupl9) involved in the regulation of glucose and lipid metabolism (Zhou and Rui, 2010) . The effect of the *AA diet in LERKO mice was very different, for which the extent of the response was very limited, not only considering the number of genes involved, but also in terms of significance (-Logiopadj) and intensity of the effect (Log2FC) (Figure 6e) .

In ERa ^f/f SHAM females (Figures 6f-g) , the *AA diet mainly influenced the expression of the genes connected to ATP synthesis and metabolism and the mitochondrial respiratory chain (Figure 7a) , the assembly pathway (Figure 7b) , the cell cycle (Figure 7c) and the transcription regulation (Figure 7d) of the p-catenin TCF. Unlike ERa ^f/f , the effect of the *AA diet in LERKO mice does not lead to the increase in genes relevant to mitochondrial respiration and for transcription, but appears to be restricted to DNA repair (Figure 6h) and to the regulation of extracellular structural proteins (Figure 6i) .

The identity analysis of the DEGs revealed that in LERKO mice fed the *AA diet the regulated biological processes are compatible with an amino acid-dependent activation of the TORC1 complex, showing an increase in the DNA repair mechanisms (Figure 7e) , a decrease in the TGFp signalling pathway (Figure 7f) (Osman et al., 2009) and the ability to induce genes for collagen deposition (Figure 7g) and extracellular matrix organisation (Figure 7h) .

These results suggest that, in the absence of ERa, the response of the liver to the *AA diet is compatible with the expected activation of the TORC1 complex (as occurs in many cells, including extra-hepatic cells) , but when the receptor is present the effect of the *AA diet is more compatible with an activation of the ERa transcriptional activity, resulting in a reflection on the metabolic pathways known to be under the control of ERa.

This led to the analysis of the role played by the hepatic ERa in mediating the effects of the *AA diet after ovariectomy by comparing the ERa ^f/f mice and LERKO mice. The analysis was performed by including all DEGs identified in the previous analysis, thus a total of 1305 genes.

In ERa ^f/f females, ovariectomy had clear consequences which were partly compensated by the *AA diet. The ablation of hepatic ERa per se had consequences for the hepatic transcriptome that do not appear to be dramatic (in fact, the number of DEGs is 107 compared to ERa ^f/f females) . The effect of ovariectomy in LERKO mice has a greater effect, but the DEGs compared to LERKO SHAM mice are only 174, therefore, in the liver of LERKO mice, ovariectomy does not disrupt the transcriptome as much as in ERa ^f/f mice, for which 912 DEGs were detected.

More relevant, however, is the fact that the effect of the *AA diet in LERKO OVX mice is clearly different than in ERa ^f/f OVX. The *AA diet failed to realign the hepatic gene expression profile to that of the ERa ^f/f mice (Figure 8) . In fact, in the LERKO OVX mice fed the *AA diet the liver transcriptomic profile is almost superimposable (~97%) to that of ERa ^f/f OVX fed the CRTL diet demonstrating that, unlike ERa ^f/f mice, the *AA diet is not able to oppose the metabolic disorders induced by ovariectomy, thus causing lipid deposit in the liver and body weight gain.

Thus, these results provide a further indication of the relevance of hepatic ERa as a sensor and modulator of hepatic metabolism.

Defeminization of the liver metabolic profile after ovariectomy

As mentioned above, the liver is a sexually dimorphic organ and it has been proposed that the lower incidence of metabolic and cardiovascular diseases that characterises fertile females is connected to the tight overall control that the female liver exerts over energy metabolism (Della Torre and Maggi, 2017; Maggi and Della Torre, 2018) . The fact that, after the cessation of the ovarian functions, females rapidly lose their metabolic advantage over males (Della Torre et al., 2014) suggests that circulating oestrogens play an important role in maintaining sexual dimorphism in the liver (Della Torre and Maggi, 2017; Maggi and Della Torre, 2018) .

To better understand the role of circulating oestrogens and of the ovarian activity in maintaining this dimorphism, genes differentially expressed in the liver of fertile females and ovariectomised females were compared with genes differentially expressed in the liver of females and males. Figure 9a shows that globally, with the stringency of the analysis used above, about 25% of the more than 900 genes responding to ovariectomy are sexually dimorphic (Della Torre et al., 2018) ; the liver content of most of them became identical to the male content after ovariectomy. This may support the idea that circulating oestrogens play an important role in maintaining hepatic sexual dimorphism.

The vast majority of genes that following ovariectomy are in the liver of females at levels comparable to those of males are concentrated in three families: Cyp450, Mups and Sult (Figure 9 b-d) .

Cyp and Sult are gene families mostly involved in steroid detoxification and catabolism, while Mups preside over glucose and lipid metabolism.

This indicates that, along with an easier control over lipid metabolism, the male liver has a lower detoxifying capacity than that of fertile females: this could explain the different susceptibility of males and females to liver damage and cardiovascular consequences and why females, after the cessation of ovarian functions , become more susceptible to liver and cardiovascular disorders .

The values of each amino acid in the composition of the invention, calculated as mg/ kg/day, were obtained starting from the study carried out , taking into account the di f ferent energy metabolism of mice and humans and the daily requirement of the single amino acids recommended by the

FAO .